vendor-connectors 1.1.2

Universal vendor connectors with transparent secret management, including Cursor AI agents, Anthropic Claude, and Meshy AI for 3D assets

Vendor Connectors

Universal vendor connectors with transparent secret management.

Vendor Connectors provides standardized access to cloud providers, third-party services, and AI APIs. Each connector offers three interfaces: a direct Python API, LangChain-compatible tools for AI agents, and MCP servers for Claude Desktop and similar clients.

Connectors

AI/Agent

• Anthropic - Claude AI message generation, token counting, and model selection
• Cursor - Cursor Background Agent API for AI coding agent management

Cloud Providers

• AWS - Boto3-based client with role assumption and retry logic
• Google Cloud - Workspace and Cloud Platform APIs with lazy credential loading

Services

• GitHub - Repository management, GraphQL queries, and file operations
• Slack - Bot and app integrations with rate limiting
• Vault - HashiCorp Vault with Token and AppRole auth
• Zoom - Meeting and user management
• Meshy - AI 3D asset generation (text-to-3D, rigging, animation, retexture)

---

Installation

pip install vendor-connectors

Optional Extras

pip install vendor-connectors[langchain]     # LangChain / LangGraph tool adapters
pip install vendor-connectors[crewai]        # CrewAI tool adapters
pip install vendor-connectors[meshy,mcp]     # Meshy MCP server support
pip install vendor-connectors[webhooks]      # Meshy webhooks
pip install vendor-connectors[all]           # Everything

The crewai extra currently resolves to the first patched CrewAI line, 1.14.2rc1+, until upstream ships the same dependency floor in a stable release.

Quick Start

Using VendorConnectors (Recommended)

from vendor_connectors import VendorConnectors

vc = VendorConnectors()

# Cloud providers
s3 = vc.get_aws_client("s3")
google = vc.get_google_client()

# Services
github = vc.get_github_client(github_owner="myorg")
slack = vc.get_slack_client()
vault = vc.get_vault_client()

# AI
claude = vc.get_anthropic_client()
cursor = vc.get_cursor_client()

Using Individual Connectors

from vendor_connectors import AWSConnector, GithubConnector

aws = AWSConnector(execution_role_arn="arn:aws:iam::123456789012:role/MyRole")
s3 = aws.get_aws_client("s3")

github = GithubConnector(
    github_owner="myorg",
    github_repo="myrepo",
    github_token=os.getenv("GITHUB_TOKEN")
)

Three Interfaces Per Connector

Every connector follows the same pattern:

# 1. Direct Python API
from vendor_connectors.meshy import text3d

# 2. LangChain Tools (works with LangChain, CrewAI, LangGraph)
from vendor_connectors.meshy.tools import get_tools

# 3. MCP Server (for Claude Desktop, Cline, etc.)
from vendor_connectors.meshy.mcp import run_server

Environment Variables

Variable	Description
ANTHROPIC_API_KEY	Anthropic Claude API key
CURSOR_API_KEY	Cursor Background Agent API key
AWS_*	Standard AWS credentials
EXECUTION_ROLE_ARN	AWS role to assume
GITHUB_TOKEN	GitHub personal access token
GOOGLE_SERVICE_ACCOUNT	Google service account JSON
SLACK_TOKEN / SLACK_BOT_TOKEN	Slack tokens
VAULT_ADDR / VAULT_TOKEN	Vault connection
MESHY_API_KEY	Meshy AI API key

---

Contributing

Contributions are welcome! Please see the Contributing Guidelines for more information.

Project Links

• PyPI
• GitHub
• Documentation
• Changelog